Ivosidenib with venetoclax ± azacitidine in IDH1-mutated myeloid malignancies

Curtis Lachowiez from The University of Texas, MD Anderson Cancer Center, Houston, TX, US, reported an interim analysis at the 2020 American Society of Clinical Oncology (ASCO) Virtual Meeting, from a trial evaluating a new combination of venetoclax (VEN) with ivosidenib (IVO) in patients with isocitrate dehydrogenase-1 gene (IDH1)-mutated hematologic malignancies (NCT03471260).¹

As a single-agent, IVO, an IDH1-mutant inhibitor, demonstrated a 42% complete response (CR) rate in treatment naïve patients,² and 30% CR amongst patients with relapsed/refractory (R/R) acute myeloid leukemia (AML).³ These results led to the approval of IVO for both settings by the U.S. Food and Drug Administration (FDA) in 2019 and 2018, respectively.

The increased reliance on BCL-2 signaling observed in samples from patients with IDH1-mutated AML (present in 7–14% of cases) was the rationale behind the combination of IVO with VEN, a BCL-2 inhibitor. Below is summarized the first results of the IVO + VEN combination with or without azacitidine (AZA), a hypomethylating agent, in IDH1-mutated myeloid malignancies.

Study design¹

The key objectives of the study were to evaluate IVO's safety and tolerability in combination with VEN ± AZA, and to determine the maximum tolerated dose and recommended dose for phase II trials. Efficacy was also explored in terms of overall response rate, including measurable residual disease (MRD) assessment by flow cytometry, and time to event endpoints. 

The phase Ib or dose-escalation part of the study here reported, established four different cohorts to test the administration of 500 mg of IVO, once daily, with varying doses of VEN (400 mg or 800 mg daily) ± 75 mg/m², days 1–7, of AZA.

Pharmacokinetics studies showed that VEN’s metabolism is faster when administered with IVO, leading to a 53% decrease in its blood concentration compared with its availability when prescribed as single-agent. Therefore, this study included a cohort with a higher dose of VEN (800 mg) than the approved dose, when administered with hypomethylating agents (400 mg).

For the future phase II or efficacy part of the study, the investigators will go forward with the dosing schedule selected from phase Ib, and expect to include 40 patients: Half treatment naïve and half R/R.

Results¹

To date, 21 patients with IDH1 R132 mutated AML have been enrolled in Cohorts 1–3. See Table 1 for patients’ baseline characteristics.

Table 1. Patient baseline characteristics

Responses were fast in the bone marrow and correlated with higher blood cell counts on peripheral blood. Moreover, the numbers improved with every additional cycle administered. All patients who achieved MRD negativity by flow cytometry presented better outcomes than those who remained MRD-positive.  The summary of response rates in every cohort and by subgroups can be found in Table 2.

The overexpression of BCL-2 is known as one mechanism of resistance to hypomethylating agents; therefore, the reported rates of complete response and MRD negativity in Cohort 3 might indicate a synergistic mechanism of action within the triplet combination. However, rates of myelosuppression and thrombocytopenia were also significantly higher in this cohort. These could be caused by the myelotoxicity added by AZA, or the poor baseline condition of the patient's bone marrow from this cohort (six/eight patients had received prior chemotherapy). The authors highlighted the careful consideration of the results due to the small sample size and a short follow-up.

After a median follow-up of 7 months, the median overall survival was not reached, and the median event-free survival was 9.4 months. See Table 2 for the results of survival outcomes.

Table 2. Response rates and survival outcomes reported by cohort and disease

The investigators analyzed patients with durable responses and observed that active signaling mutations (e.g., RAS, receptor tyrosine kinase pathway or phosphatase pathway mutations), previously reported as mutations associated to resistance to IVO and VEN, were present in 18% of patients compared with 66% of patients with no response or relapse. Although there was no significant correlation between the absence of these mutations and CRc achievement, this might indicate a higher probability of achieving MRD negativity. The median duration of response was also significantly higher in patients with no active signaling mutations versus patients with these mutations (11.9 months vs 1.6 months; p value = 0.043).

Across all cohorts, patients who achieved CR and MRD negativity by flow cytometry have not relapsed and, to date, remain event-free after a median follow-up of 11 months. More patients from Cohort 3 achieved MRD negativity, probably because of the addition of the third agent, AZA.

Safety

Common Grade 1–2 adverse events (AEs), experienced by 10% or more patients, included diarrhea, nausea, and vomiting. While common Grade 3–4 AEs included pneumonia and febrile neutropenia, which were considered as expected events for patients with advanced hematological malignancies.¹

AEs of special interest were tumor lysis syndrome, in two patients, and IDH differentiation syndrome (DS), in four patients. All managed with medication alone and resolved. One death occurred during the study.¹

DS is defined by the occurrence of two or more defined symptoms, i.e., dyspnea, fever, weight gain, unexplained hypotension, acute renal failure, or pleuropericardial effusion. Patients with two or three criteria are classified as moderate DS, and those with at least four, are classified as severe DS.

After initial studies with IDH1 and IDH2 inhibitors, DS was identified as an AE of special interest being experienced by at least 11–14% of patients. Due to suspected underreporting in these trials, the FDA recently analyzed the incidence of this AE in 393 patients with R/R IDH-mutated AML treated with IVO or enasidenib (ENA, an IDH2-mutated inhibitor), and published its results in Clinical Cancer Research.⁴

An algorithm, together with an FDA expert reviewer, was able to detect DS symptoms in 19% of patients receiving IVO or ENA. When taking into account the baseline characteristics of these patients, they observed that those who presented higher percentages of bone marrow blast and peripheral blast before treatment or were diagnosed with secondary AML, had a higher risk of experiencing IDH-inhibitor-associated DS (IDH-DS) during treatment.

Most confirmed cases of IDH-DS were moderate in severity, and fatalities were rare, although more than half were categorized as Grade ≥ 3 AEs. The median time to onset was > 7 days following the initiation of IDH inhibitor therapy. Additionally, for 12% and 15% of patients treated with IVO and ENA, respectively, it was not an isolated event and experienced at least one repetition of it, with a median of 19 and 22.5 days between episodes. Moreover, patients suffering at least one episode of IDH-DS had a shorter exposure to the drug, potentially leading to worst outcomes in terms of CR, median duration of response, and median overall survival (4.7 vs 10.0 with IVO, 8.0 vs 8.3 with ENA), compared with those who did not experience any DS symptoms.

A total of 59% and 56% of patients with IDH-DS treated with IVO and ENA, respectively, received systemic corticosteroids for DS's signs or symptoms. Hydroxyurea and furosemide were also more frequently used in patients with DS compared with those without DS. The use of leukapheresis, mechanical ventilation, and dialysis was rare (< 5%) in the whole cohort.

From this analysis, the FDA concluded that it is of the utmost importance to standardize a diagnostic algorithm to help earlier recognition and treatment of IDH-DS in both clinical trials and clinical practice, since it is a common and potentially fatal AE of IDH-inhibitors therapy.

Conclusions¹

The results reported from this interim analysis demonstrate that IVO + VEN ± AZA is an active, molecularly targeted, well-tolerated regimen for advanced IDH1-mutated myeloid malignancies.

Overall, the new combination leads to high complete remission rates in a clinically adverse patient population and half of which were also MRD-negative. However, response assessment is ongoing, and Cohort 4, with a higher dose of VEN, is recruiting participants. Therefore, a longer follow-up is needed to establish the recommended dose for phase II and to further assess the efficacy of this regimen.

Review the most relevant results of IVO and ENA as single-agents and in combination for patients with AML in this previous article from the AML Hub.